AI Prompts

Design globally resilient event ingestion with clear consistency and failover semantics.

Design a reliable batch or streaming pipeline from source to warehouse, including load pattern, idempotency, and failure handling.

Draft an architecture decision record with real options, consequences, and reversal cost so the team can commit without burying disagreement in chat.

Write a decision memo among prompting, RAG, and fine-tuning with data needs, ops cost, and a reversal path instead of a default to fine-tune.

Design context packing, memory, and summarization so long sessions stay correct, with eviction rules and evals for silent loss of constraints.